Deep convolutional neural networks (CNNs) have emerged as a new paradigm for Mammogram diagnosis. Contemporary CNN-based computer-aided-diagnosis systems (CADs) for breast cancer directly extract latent features from input mammogram image and ignore the importance of morphological features. In this paper, we introduce a novel end-to-end deep learning framework for mammogram image processing, which computes mass segmentation and simultaneously predicts diagnosis results. Specifically, our method is constructed in a dual-path architecture that solves the mapping in a dual-problem manner, with an additional consideration of important shape and boundary knowledge. One path, called the Locality Preserving Learner (LPL), is devoted to hierarchically extracting and exploiting intrinsic features of the input. Whereas the other path, called the Conditional Graph Learner (CGL), focuses on generating geometrical features via modeling pixel-wise image to mask correlations. By integrating the two learners, both the cancer semantics and cancer representations are well learned, and the component learning paths in return complement each other, contributing an improvement to the mass segmentation and cancer classification problem at the same time. In addition, by integrating an automatic detection set-up, the DualCoreNet achieves fully automatic breast cancer diagnosis practically. Experimental results show that in benchmark DDSM dataset, DualCoreNet has outperformed other related works in both segmentation and classification tasks, achieving 92.27% DI coefficient and 0.85 AUC score. In another benchmark INbreast dataset, DualCoreNet achieves the best mammography segmentation (93.69% DI coefficient) and competitive classification performance (0.93 AUC score).

In this paper, we investigate the mean communication rate loss (MCRL) due to beam cusping with respect to multiple input multiple output (MIMO) spatial dimensions, beamforming codebook sizes, and beam scan spaces for beamforming operations controlled by the medium access control (MAC) layer beamforming protocol. The intractability of the analytical investigation for a closed-form expression leads to the usage of Monte Carlo numerical experiments. For discrete Fourier transform (DFT) beamforming codebooks, the MCRL increases with an increase in the MIMO dimensions while it decreases with an increase in the beam scan space from the array broadside. The MCRL becomes negligible as the number of beams for selection is increased to four times the number of DFT beams. These results will be helpful for determination of MIMO configuration, angular coverage zone, codebook size, and phase-shifter resolution for a minimal MCRL. This is especially relevant for sparse MIMO channels of millimetre-wave (mmWave) and terahertz (THz) wireless systems in the fifth generation (5G) and beyond mobile networks.

The millimetre-wave (mmWave) band communication system suffers from its susceptibility to blockage. Unlike human blockage, which has been extensively investigated in the literature, the impact of other small objects on signal strength, such as metal road signs, is not completely understood. In this study, we measure and investigate the attenuation of different types of relatively small objects. It is found that the resulting loss is in the range of 15 to 30 dB depending on the blocker size and type. We have investigated the applicability of the existing 3rd generation partnership project (3GPP) blockage model for small-size blockers, with and without including the antenna gain. Our results suggest that the 3GPP model underestimates the attenuation caused by small-size blockers, and including the antenna gain into the 3GPP model does not show a significant improvement. The multiple knife-edge (MKE) model shows a better performance in capturing the measured attenuation with a smaller error.

Wireless Local Area Networks (WLANs) using the standard Distributed Coordination Function (DCF) have widely been deployed. A large number of wireless multiple access techniques were studied for achieving efficient and fair WLANs; however, little attention has been paid to access methods for improving total network throughput performance while maintaining fairness in the presence of legacy devices using DCF. So, we propose a novel random access technique called polite when unsynchronized (PWU) for seamless transition to efficient and fair WLANs. In PWU, each node has a new parameter, called the politeness level, which is a number between zero and one. A node with a higher politeness level becomes polite more frequently, and it tends to sense the channel for a bit longer compared to nodes usinga DCF. A politeness control algorithm (PCA) is studied, and it is shown that nodes using PCA maintains a high level of fairness to nodes using DCF in terms of channel access opportunities when they coexist.

We present, for the first time, a novel deep neural network architecture called DualCoreNet with a dual-path connection between the input image and output class label for mammogram image processing. This architecture is built upon U-Net, which non-linearly maps the input data into a deep latent space. One path of the DualCoreNet, the locality preserving learner, is devoted to hierarchically extracting and exploiting intrinsic features of the input, while the other path, called the conditional graph learner, focuses on modeling the input-mask correlations. The learned mask is further used to improve classification results, and the two learning paths complement each other. By integrating the two learners our new architecture provides a simple but effective way to jointly learn the segmentation and predict the class label. Benefiting from the powerful expressive capacity of deep neural networks a more discriminative representation can be learned, in which both the semantics and structure are well preserved. Experimental results show that DualCoreNet achieves the best mammography segmentation and classification simultaneously, outperforming recent state-of-the-art models.

For decades wireless local area networks (WLANs) using the distributed coordination function (DCF) protocol have been suffering from packet collisions causing throughput performance degradation. In order to resolve this issue, we propose a novel resource scheduling scheme, named 3 bits of reservation information (3bRI). In 3bRI, the scheduling information of the next three time slots is broadcast in ACK frames, and the information can be overheard by neighboring stations. Each active neighboring station checks the overheard scheduling information before starting a transmission, and defers its transmission to avoid a packet collision when the time slot has already been reserved. Simulation results show that the proposed method improves the total throughput performance by 55% compared to DCF, and 36% compared to centralized random backoff (CRB) when the number of active nodes is 40. It also maintains a high level of fairness to legacy devices using DCF when devices using 3bRI coexist with legacy devices.

Investigating the blockage effects on signal strength for the mmWave band used in vehicle-to- infrastructure (V2I) communications is the main goal of this study. The proposed Markov chain model successfully captures the dynamics of the channel caused by blockers surrounding a moving vehicle. The transition probabilities are calculated both analytically and via a geometric simulation, which are found to match well. These transition probabilities are then used to build a Markov chain model. The proposed Markov model accurately captures the autocorrelation (ACF) and the empirical cumulative distribution function (ECDF) of the link capacity for a moving vehicle, matching the results of the geometric model. Although the Markov chain requires statistical stationarity to work well, this paper proposes a way to switch between several Markov chains to tackle the scenarios with non-stationary blocker densities such as a straight-line road.

Wi-Fi devices operate following the 802.11 distributed coordination function in order to fairly use the channel that the devices share. However, the throughput performance of the Wi-Fi networks is known to be degraded due to packet collisions. So, we propose a novel multiple access protocol, called centralized random backoff (CRB) for collision-free Wi-Fi networks. In CRB, after a successful reception of a data frame from a station, the access point allocates a unique backoff state to the station by means of the ACK frame. We evaluate its performance by comparing to that of a deterministic backoff mechanism. Evaluation results show that CRB significantly improves the throughput performance by reducing collisions, and it allows a larger number of nodes to operate in a collision-free state without dynamic parameter adjustment.

Recent research has shown the feasibility of in-band full duplex (FD) wireless operation allowing wireless nodes to send and receive in the same frequency band at the same time. In this paper, we propose a novel FD medium access control (MAC) protocol, and discuss two issues related to the FD MAC layer. These include fairness between half duplex (HD) and FD users, and a channel access mechanism called centralized random backoff (CRB) to resolve collisions caused by the random backoff algorithm adopted in the IEEE 802.11 MAC standard. Simulation results show that the protocol provides evenly distributed channel access opportunities to both FD and legacy HD users. Moreover, it is demonstrated by simulation that using CRB complements the random backoff algorithm and improves the performance of HD and FD MAC protocols for Wi-Fi networks.

In this paper, a comprehensive power performance analysis of a novel Adaptive Switching Algorithm for an iterative-MIMO system is investigated with the prime goal of minimizing energy consumption in the receiver. The algorithm works by switching between a high performance detection method, the Fixed Sphere Decoding, and a much lower complexity algorithm, the Vertical-Bell Laboratories Layered Space-Time Zero Forcing technique, controlled by a threshold according to the mutual information calculated during each transmission. Results show significant improvements over current non-adaptive receivers, where energy savings of more than 60% can be obtained using on the latest Xilinx®Virtex-7 FPGA hardware.

This paper proposes a novel three-dimensional (3D) theoretical regular-shaped geometry-based stochastic model (RS-GBSM) and the corresponding sum-of-sinusoids (SoS) simulation model for non-isotropic multiple-input multiple-output (MIMO) vehicle-to-vehicle (V2V) Ricean fading channels. The proposed RS-GBSM, combining line-of-sight (LoS) components, a two-sphere model, and an elliptic-cylinder model, has the ability to study the impact of the vehicular traffic density (VTD) on channel statistics, and jointly considers the azimuth and elevation angles by using the von Mises Fisher distribution. Moreover, a novel parameter computation method is proposed for jointly calculating the azimuth and elevation angles in the SoS channel simulator. Based on the proposed 3D theoretical RS-GBSM and its SoS simulation model, statistical properties are derived and thoroughly investigated. The impact of the elevation angle in the 3D model on key statistical properties is investigated by comparing with those of the corresponding two-dimensional (2D) model. It is demonstrated that the 3D model is more accurate to characterize real V2V channels, in particular for pico cell scenarios. Finally, close agreement is achieved between the theoretical model, SoS simulation model, and simulation results, demonstrating the utility of the proposed models.

In this work, we investigate a general performance evaluation framework for network selection strategies (NSSs) that are used in 3G-WLAN interworking networks. Instead of simulation, this framework is based on models of NSSs and is constructed using a stochastic process algebra, named Performance Evaluation Process Algebra (PEPA). It captures the traffic and mobility characteristics of mobile nodes in 3G-WLAN interworking networks and has a good expression of the behavior of the mobile nodes using different NSSs. Commonly used NSSs are evaluated from the perspectives of average throughput, handover rate, and network blocking probability. Results of the evaluation explore the effect of these NSSs on both mobile nodes and networks, as well as their characteristics in different mobility and traffic scenarios.

A significant challenge of cognitive radio (CR) is to perform wideband spectrum sensing in a fading environment. In this paper, a novel multi-rate sub-Nyquist spectrum detection(MSSD) system is introduced for cooperative wideband spectrum sensing in a distributed CR network. Using only a few sub- Nyquist samples, MSSD is able to sense the wideband spectrum without full spectrum recovery. Specifically, given the low spectral occupancy, sub-Nyquist sampling is performed in each sampling channel and a test statistic is formed by using sub-Nyquist samples from multiple sampling channels. Furthermore, the use of different sub-Nyquist sampling rates is proposed to improve the system detection performance, and the performance of MSSD over both non-fading and Rayleigh fading channels is analyzed. Numerical results show that MSSD can considerably improve the wideband spectrum sensing performance in a fading scenario, with a relatively low implementation complexity and a low computational complexity.

Abstract For the practical simulation and performance evaluation of mobile‐to‐mobile (M2M) communication systems, it is desirable to develop accurate M2M channel simulation models for more realistic scenarios of non‐isotropic scattering. In this paper, by using a ‘double‐ring’ concept to describe M2M non‐isotropic scattering environments, we propose new deterministic and stochastic sum‐of‐sinusoids (SoS) based simulation models. The proposed simulation models extensively consider the distributions of the angle of arrival (AoA) and the angle of departure (AoD), and thus provide a good approximation to the desired statistical properties of the reference model. Copyright © 2009 John Wiley & Sons, Ltd.

This paper proposes a novel three-dimensional (3D) two-sphere regular-shaped geometry-based stochastic model (RS-GBSM) with only double-bounced rays for non-isotropic scattering narrowband multiple-input multiple-output (MIMO) mobile-to-mobile (M2M) channels. The proposed 3D model has the ability to investigate the joint impact of both the azimuth angle and elevation angle on channel statistics. Based on the proposed model, the space-time (ST) correlation function (CF) is derived and the impact of some important parameters on the resulting ST CF is investigated. Numerical results show that the 3D model results in lower ST correlations than the corresponding 2D model.

With the increasing diversity of content as well as user preferences, and the heterogeneity of devices and network technologies, content adaptation has been widely acknowledged as an effective strategy to deliver services and content to users in a variety of contexts. This paper presents an evaluation of the response time of large scale content adaptation systems being developed under the auspices of the Mobile VCE, using the high-level modelling formalism -performance evaluation process algebra (PEPA). The relevant factors of the system performance, including the operation speed of individual entities, loading and resource conditions of the system, are determined and analysed.

This paper proposes a novel three-dimensional (3D) regular-shaped geometry-based stochastic model (RS-GBSM) for non-isotropic multiple-input multiple-output (MIMO) mobile-to-mobile (M2M) Ricean fading channels. The proposed model, combining a two-sphere model and an elliptic-cylinder model, is the first 3D RS-GBSM that has the ability to investigate the impact of the vehicular traffic density (VTD) on channel statistics and jointly consider the azimuth angle and elevation angle. From the proposed model, the space-time (ST) correlation function (CF) and the corresponding space-Doppler (SD) power spectral density (PSD) are derived. Finally, some numerical results and interesting observations are given.

Energy efficiency is becoming a very important issue in modern wireless communication systems. At the same time, maintaining QoS is also critical when trying to reduce power consumption. This paper describes a heterogeneous wireless communication network architecture, and analyses its power consumption and QoS support levels. The result shows that combining cellular communications with femtocells can significantly reduce overall power consumption, depending on the uptake of femtocell usage. Also, QoS remains high when the power consumption drops. The simulation results also suggest that the femtocell implementation rate should be limited under a certain degree, around 60%, in consideration of power efficiency. Finally, indifference curves of power consumption and QoS indicate the optimal femtocell adoption schemes under different requirements.

In this paper, based on the tapped delay line (TDL) structure, we propose a geometry-based stochastic model (GBSM) for wideband multiple-input multiple-output (MIMO) mobile-to-mobile (M2M) Ricean fading channels. The proposed wideband model is the first GBSM that has the ability to study the impact of the vehicular traffic density (VTD) on channel statistics for different time delays, i.e., for every tap in our model. From the proposed model, the space-time (ST) correlation function (CF) and the corresponding space-Doppler (SD) power spectral density (PSD) are derived. Excellent agreement is achieved between the theoretical Doppler PSDs and measured data, demonstrating the utility of the proposed model.

In this paper, based on the concept of the tapped delay line (TDL) structure, we first propose a new wideband multiple-ring multiple input multiple output (MIMO) channel reference model for macro-cell scenarios. It is shown that the proposed model can easily match any given or measured power delay profile (PDP) and has the ability to jointly consider the angle of arrival (AoA), angle of departure (AoD), and time of arrival (ToA). From the proposed model, we derive the closed-form expression of the 3-dimensional (3-D) space-time-frequency (STF) correlation function (CF) for each tap, considering the interaction of the temporal, spatial, and frequency correlations. Finally, based on the proposed reference model, a deterministic simulation model is then proposed.

This paper develops a generic geometry-based stochastic model for mobile-to-mobile (M2M) Ricean fading channels. From the generic model, the level crossing rate (LCR) and average fade duration (AFD) are derived. Based on the derived expressions, we for the first time investigate the LCR and AFD for M2M channels with different vehicular traffic densities (VTDs). Excellent agreement is achieved between the theoretical results and measured data, demonstrating the utility of the proposed model.

Generative models are created to be used in the design and performance assessment of high layer wireless communication protocols and some error control strategies. Generative models can replace real digital wireless channels to significantly reduce the time and complexity of system simulation. The errors occurring in digital wireless channels are not independent but form clusters or bursts. Generative models have to produce error sequences having similar burst error statistics to those of original error sequences obtained from real digital systems. In this paper, we propose a generative hidden Markov model (HMM) with three layers. It is shown that the proposed three layered HMM can generate error sequences that have statistics compatible with those of original error sequences derived from an enhanced general packet radio service (EGPRS) transmission system.

For realistic simulation of multiple-input multiple -output (MIMO) channels under more realistic scenario of non-isotropic scattering, we propose new deterministic and stochastic sum-of-sinusoids (SoS) based simulation models. The proposed models comprehensively consider the distributions of the angle of arrival (AoA), and thus show a good approximation to the desired statistical properties of the reference model.

To simulate mobile-to-mobile (M2M) Rayleigh fading channels under more realistic scenario of non-isotropic scattering, we propose one deterministic and one stochastic sum-of-sinusoids (SoS) based simulation models. The proposed models extensively consider the distributions of the angle of arrival (AoA) and the angle of departure (AoD), and thus show a good approximation to the desired statistical properties of the reference model.

Soft decision decoding is currently used in many wireless communications systems because of its capability for increasing the coding gain. Soft generative models produce soft error sequences in order to help in the design and evaluation of the performance of many coding schemes and high layer wireless communication protocols. Errors in wireless fading channels are not independent but appear in bursts or clusters. In this paper, we design a soft Markovian based generative model with two embedded processes. The first process combines produced soft error bursts with soft error-free bursts. Whereas, in the second process new error bursts are created according to the number of soft error clusters in each soft error burst of a reference soft error sequence. The reference soft error sequence is obtained from an enhanced general packet radio service (EGPRS) transmission system. The achieved results show a satisfactory behavior compared with the reference soft error sequence.

An analytical upper bound on the ergodic capacity of Multiple-Input Multiple-Output (MIMO) systems is deduced with the aid of a simplified approach that relies on a fundamental power normalization. Given their high practical usability, we are particularly interested in dual configurations where both ends deploy two antenna elements. Contrary to the majority of related studies, where only the common case of Rayleigh fading is considered, our analysis is extended to account for Ricean fading where a deterministic Line-of-Sight (LoS) component exists in the communication link and both ends are affected by spatial correlation. In the following, it is shown that the proposed bound is applicable for any arbitrary Signal-to-Noise Ratio (SNR) and rank of the mean channel matrix. In fact, we consider both conventional and optimized MIMO configurations with equal LoS eigenvalues. Moreover, the tightness of the bound is explored where it is demonstrated that as the SNR tends to zero the bound becomes asymptotically tight; at high SNRs, the offset between empirical capacity and the bound is analytically computed which implies that an explicit asymptotic capacity expression can ultimately be obtained.

In this paper, a generic and adaptive geometrybased stochastic model (GBSM) is proposed for non-isotropic multiple-input multiple-output (MIMO) mobile-to-mobile (M2M) Ricean fading channels. The proposed model employs a combined two-ring model and ellipse model, where the received signal is constructed as a sum of the line-of-sight, single-, and doublebounced rays with different energies. This makes the model sufficiently generic and adaptable to a variety of M2M scenarios (macro-, micro-, and pico-cells). More importantly, our model is the first GBSM that has the ability to study the impact of the vehicular traffic density on channel characteristics. From the proposed model, the space-time-frequency correlation function and the corresponding space-Doppler-frequency power spectral density (PSD) of any two sub-channels are derived for a non-isotropic scattering environment. Based on the detailed investigation of correlations and PSDs, some interesting observations and useful conclusions are obtained. These observations and conclusions can be considered as a guidance for setting important parameters of our model appropriately and building up more purposeful measurement campaigns in the future. Finally, close agreement is achieved between the theoretical results and measured data, demonstrating the utility of the proposed model.

In this paper, we explore the statistical characterization of Multiple-Input Multiple-Output (MIMO) channel correlation matrices with the main focus being on their condition number statistics. More specifically, novel expressions are derived for the probability density function (PDF) and cumulative distribution function (CDF) of the MIMO condition number. Contrary to the majority of related studies, where only the common case of Rayleigh fading was considered, our investigation is extended to account for the generalized case of Ricean fading where a deterministic Line-Of-Sight (LoS) component exists in the communication link. The overall analysis is based on the principles of random matrix theory and particularly of dual complex non-central Wishart matrices; the latter represent a practical class of MIMO systems, namely dual-branch systems which are equipped with two transmit and receive antenna elements. All the derived formulae are validated through extensive simulations with the attained accuracy being remarkably good.

Spectrum sensing plays a paramount role in cognitive radio (CR), which is widely agreed to be the most promising method for alleviating the symptom of RF spectral scarcity. In this paper, we first analyze the effects of multipath/shadowing on a CR system. Furthermore, we build models for the primary user system and the CR user system separately. For the former model, we develop the possible conversation zone to represent the CR cooperative detectable zone. For the latter model, we construct a novel centralized network, called the spider-net sensing network (SNSN), to implement controllable sensing resolution and reliable sensing while avoiding harmful interference from other CR users. Finally, the numerical results show that our proposed SNSN outperforms one without a grid structure using the same combining scheme.

As a resource reservation mechanism, the Resource ReSerVation Protocol (RSVP) faces a lot of challenges when applying it to the wireless and mobile networks. The interworking problems of RSVP and mobility management protocols have been extensively discussed over the last decade. As the solutions of this problem, mobility-aware RSVP schemes that integrate RSVP and micro-mobility management are becoming more and more popular. Therefore, the investigation on how much they improve the performance of the basic RSVP is necessary and useful. Instead of the traditional simulation based approaches, in this paper we introduce a formal performance evaluation formalism, named Performance Evaluation Process Algebra (PEPA), and employ it to investigate the performance of the basic RSVP and mobility-aware RSVP. Important performance metrics such as handover blocking probability and signalling cost are presented.

Ubiquitous personal content transfer in a heterogeneous environment includes both global infrastructure based communications and local infrastructureless transfer, which leads to a hybrid networking environment. The MIP/NEMO standard can support ubiquitous content transfer, but is inefficient for local content transfer. Infrastructureless communication using ad hoc mode is often utilized by individual users to transfer local content, but it can not support device mobility. In this paper, a scheme based on a personal distributed environment (PDE) is proposed to implement ubiquitous content transfer in a hybrid networking environment. It can improve performance by combining the virtues of MIP/NEMO and the advantages of ad hoc mode.

One of the major problems of deploying RSVP in the mobile environment is called theadvanceresourcereservationproblem. If an RSVP reservation path is reserved in advance in the subnet that a mobile node will visit, the mobile node can continue its QoS session smoothly when it hands over to that subnet. However, if too many network resources are used for advance reservation, new QoS sessions originating from that subnet will experience a higher probability of being blocked needlessly. In this paper, we propose a new advance resource reservation scheme that properly constrains the amount of advance reservations in a subnet and only allows the mobile nodes with large value of session-to-mobility ratio (SMR) to make advance reservation. We evaluate the proposed scheme and the results show that our scheme can effectively reduce both active and passive reservation blocking probabilities and achieves a better utilisation of the network resources, especially when the traffic intensity is high.

In this paper, the performance of a multiple-input multiple-output (MIMO) system is assessed in the case of vehicle-to-roadside communications. We investigate a line-of- sight (LoS) scenario where a specular wavefront impinges on the receive side. Under these conditions, the channel response is usually rank deficient due to the high correlation between the spatial LoS responses, unless specific antenna geometries are employed in order to achieve subchannel orthogonality. For our investigation, a recently proposed criterion, which maximizes the LoS-channel rank and eventually the channel capacity, is revised for different array configurations.

An analytically solvable mathematical model is presented for an IEEE 802.11 network, where two terminals, both visible to an access point but not to each other, contend for the channel. This situation, known as the hidden node topology, has been identified as a key reason for the degradation of the protocol's performance. It is shown that the renewal theory, the analytical tool used traditionally, is not suitable for this topology and a discrete time Markov chain is introduced for the modelling of the channel contention. The model permits the accurate computation of several key performance metrics and is in agreement with simulation results over a variety of scenarios.

Recently research interest in the performance analysis of wireless networks was revived. An issue of an utmost importance in this class of networks, classified as one of the severest reasons for the degradation of their performance, is the hidden terminal problem. In this paper we argue that an accurate analysis of the effect of hidden nodes in the performance of a random access protocol is still an open issue. Firstly, we thoroughly explain the reasons behind the limitations of previous modelling methods, and show that their performance is reliable only for certain configurations. Secondly, and most importantly, we propose a novel method of modelling time that uses a fixed-length channel slot as the unit of time and does not rely on renewal theory. With these features the model is able to successfully take into account the desynchronisation of nodes in a hidden terminal environment. Our analytical model is shown to have a very close match to simulation results for an IEEE 802.11 MAC protocol and for all the system parameters considered, unlike conventional methods.

Abstract In this paper, the performance of a dual Multiple‐Input Multiple‐Output (MIMO) communication system is investigated using data obtained from an indoor measurement campaign. We focus on the double‐directional channel response of each frequency band and the degree of similarity between them. In general, when a Frequency Division Duplex (FDD) system is employed, as for example in Universal Mobile Telecommunications System (UMTS), it is of vital importance to study whether the measurement results from the uplink could be directly extrapolated to the downlink direction, and vice versa. By applying two conventional beamforming techniques and the well known MUltiple SIgnal Classification (MUSIC) algorithm, the angular domain performance of both bands was assessed. The initial results presented herein are promising enough as the joint spectra seem to be remarkably similar implying that the underlying propagation mechanisms at both bands are also similar. Copyright © 2007 John Wiley & Sons, Ltd.

To overcome the drawbacks of the mobile IPv6 protocol on handling local mobility management, IETF proposed the HMIPv6 protocol which introduces an intermediate mobility anchor point (MAP) to hide the movement of a mobile node within a local area. However, the MAP forms a bottleneck in the network since all the traffic destined for its served nodes has to go through it. Most research on HMIPv6 focuses on protocol optimisation, and performance analysis of HMIPv6 is usually simulation-based. In this paper, we employ a performance evaluation formalism named PEPA to investigate the performance tradeoffs of MAPs in HMIPv6. Performance measures such as response time and MAP utilisation are presented.

A promising new class of wireless networks has recently emerged, the so-called Wireless Mesh Networks (WMNs). WMNs inherit many advantages from both wired infrastructure and also ad-hoc networks. However, in spite of their distinct characteristics, they also face many technical problems that were present in older classes of wireless networks, such as wireless LANs or pure ad-hoc networks. The much discussed hidden terminal problem is one of them. We argue that an accurate analysis of the effect of hidden nodes in the performance of a random access protocol is still an open issue. In this paper we show the problems and the limitations stemming from common assumptions used in the literature and propose a novel method of modeling time that uses a fixed size channel slot as the unit of time in order to take into account the desynchronisation of nodes. Finally we apply our general method to the classic hidden terminal problem incorporating an iterative step. The proposed analytical model is shown to have a very good match to simulation results for an IEEE 802.11 MAC protocol.

The last couple of years were characterised by the revival of interest in the performance evaluation of wireless multihop networks. An issue of great significance in carrier-sense based multihop networks is the hidden node problem. In our recent work we showed that previous techniques are suitable for limited node configurations and we proposed a novel time model that relaxes most assumptions in the literature and exhibits high accuracy of performance evaluation. The present paper extends our recent work, which was based on a constant contention window assumption, to the binary exponential backoff (BEB) mechanism of IEEE 802.11, with an equally high degree of agreement to simulation results. Key points of our analysis are the consideration of different collision probabilities, dependent on the backoff stage, and the use of a discrete time Markov chain embedded in the channel state evolution that allows us to compute the 'backoff-dwelling' distribution of each transmitter

Distributed Channel-Adaptive Fair Allocation (DCAFA) is proposed to extend the IEEE 802.11e Enhanced Distributed Coordination Function (EDCF), by halving the contention window (CW) after ζ consecutive successful transmissions to reduce the collision probability when channel is busy. The scheme computes an adaptive threshold function for each priority queue by taking into account the channel load. Simulation results show that DCAFA outperforms EDCF in-terms of global throughput, throughput of traffic class, frame discard probability and average delay in medium and high load cases. They also show that the proposed analytical model is accurate and achieves a high degree of fairness among applications of same priority level.

A novel stochastic wideband dynamic spatio-temporal indoor channel model which incorporates both the spatial and temporal domain properties as well as the dynamic evolution of paths when the mobile moves is proposed based on the concept of a Markov process. The derived model is based on dynamic measurement data collected at a carrier frequency of 5.2 GHz in typical indoor environments. Multipath components are estimated using the super-resolution frequency domain space-alternating generalized expectation maximization algorithm prior to identification of path "birth" and "death" using a new data analysis method. Analysis shows that multiple births and deaths are possible at any instant of time. Furthermore, correlation exists between the number of births and deaths. Thus, an M-step 4-state Markov channel model (MCM) is proposed in order to account for these two effects. The spatio-temporal variations of paths within their lifespans are taken into consideration by the spatio-temporal vector which was found to be well modeled by a Gaussian probability density function while the power variation can be modeled by a simple low-pass filter. In addition, the methodology used to extract the MCM parameters from the measurement data is also presented. Due to the distinction in the birth-death statistics, the model is generalized through segmentation of the measurement runs and can be completely parameterized by several sets of Markov parameters associated with the type of environment and scenario under consideration. The implementation of the model is also detailed and, finally, the model is evaluated by comparing key statistics of the simulation results with the measurement results.

A novel stochastic wideband dynamic directional indoor channel model which incorporates both the spatial and temporal domain properties as well as the dynamic evolution of paths when the mobile moves is proposed based on the concept of a Markov process. The derived model is based on dynamic measurement data collected at a carrier frequency of 5.2 GHz in several typical indoor environments. Analysis shows that multiple births and deaths are possible at any instant of time. Furthermore, correlation exists between the number of births and deaths. Thus, an M-step, 4-state Markov channel model is proposed in order to account for these two effects. The spatio-temporal variations of paths within their lifespan are taken into consideration by the spatio-temporal vector which was found to be well-modeled by a Gaussian probability density function while the power variation can be modeled by a simple low pass filter. The implementation of the model is detailed and finally, the model validity is evaluated by comparing key statistics of the simulation results with the measurement results.

In this paper, the spatial and temporal dispersion and correlation properties of a wideband dynamic directional indoor channel at 5.2 GHz are presented based on extensive measurement campaigns in four different environments under various propagation scenarios. Channel spatio-temporal dispersions are assessed in terms of the rms delay spread (DS) and rms azimuth spread (AS), respectively, while the correlation properties are determined by the spatio-temporal correlation coefficient. A strong correlation is observed with the presence of the line-of sight (LOS) path but degraded when the LOS is obstructed. The variations of the spatio-temporal dispersions with mobility are also investigated where higher values of spatio-temporal dispersions were found when the transmitter and receiver separation increased particularly in a more cluttered environment. The average values of the rms DS and rms AS for all investigated environments and scenarios are also given. These parameters can be used as empirical values for the 5 GHz band WLAN systems.

In this paper, a new statistical wideband indoor channel model which incorporates both the clustering of multipath components (MPCs) and the correlation between the spatial and temporal domains is proposed. The model is derived based on measurement data collected at a carrier frequency of 5.2 GHz in three different indoor scenarios and is suitable for performance analysis of HIPERLAN/2 and IEEE 802.11a systems that employ smart antenna architectures. MPC parameters are estimated using the super-resolution frequency domain space-alternating generalized expectation maximization (FD-SAGE) algorithm and clusters are identified in the spatio-temporal domain by a nonparametric density estimation procedure. The description of the clustering observed within the channel relies on two classes of parameters, namely, intercluster and intracluster parameters which characterize the cluster and MPC, respectively. All parameters are described by a set of empirical probability density functions (pdfs) derived from the measured data. The correlation properties are incorporated in two joint pdfs for cluster and MPC positions, respectively. The clustering effect also gives rise to two classes of channel power density spectra (PDS)-intercluster and intracluster PDS-which are shown to exhibit exponential and Laplacian functions in the delay and angular domains, respectively. Finally, the model validity is confirmed by comparison with two existing models reported in the literature.

This paper considers the impact of interleaving and combining multiple DMT symbols in a single RS codeword on the performance of Asymmetrical Digital Subscriber Lines (ADSL) in the presence of impulse noise. In recent years there has been an increasing interest in xDSL technology due to its high data rate capability and relative ease of deployment. Impulse noise is one of the main impairments for xDSL systems and different framing techniques are aimed at mitigating the impact of impulse noise. In the ADSL standard techniques for interleaving and combining of several consecutive symbols into a single FEC codeword are used. This paper considers the performance of the above techniques at higher layers of the protocol stack. The results are based on simulations of a generic ADSL system with a new broadband-oriented impulse noise model. The conclusions include a comparison between the packet error rates of the two techniques at different bit rates and the interaction between them when used together. The aim of this research is to find a suitable set of framing parameters and help designers in their choice of single or dual latency ADSL systems.

This paper considers the performance of Digital Subscriber Lines (DSL) in the presence of impulse noise in terms of data errors. Impulse noise is one of the biggest impairments for xDSL systems. An analysis of error statistics of transmitted data for two popular versions of xDSL - ADSL and SHDSL - is reported here. On the basis of a recent novel impulse noise model a model of the impact of impulse noise on higher protocol levels has been developed. Specifics of framing for the different DSL versions have also been taken into account. Simulation results for statistics of data error due to impulse noise based on the new byte level models are presented.

In this paper, the serial interference cancellation (SIC) technique and the frequency domain SAGE (FD-SAGE) algorithm are jointly used to detect and estimate the radio channel parameters of interest. The implementation of the SAGE algorithm in the frequency domain is novel. Furthermore, the parallel interference cancellation (PIC) technique in the standard SAGE algorithm is replaced by the SIC technique. The SIC technique demonstrates more stable performance especially in a multipath rich environment. The two-dimensional (2-D) FD-SAGE algorithm and the SIC technique are introduced and their performance demonstrated by using real indoor channel measurement data to jointly estimate the number of multipath components (MPCs), their time-of-arrivals (TOAs), angle-of-arrivals (AOAs) and complex amplitudes. Their performance is evaluated and compared using synthetic data and results using 2-D unitary ESPRIT (another form of super-resolution algorithm).

This paper considers the cell error performance of ATM over digital subscriber lines (DSL) in the presence of impulse noise. In previous years there has been an increasing interest in xDSL technology due to its relatively broad bandwidth and ease of deployment. ATM is a preferred protocol over DSL. ATM, however, has been designed for low bit error rates. Despite error correction protocols, DSL cannot always overcome bursty errors, particularly impulse noise. Interest in DSL has triggered research into impulse noise. A new model has been developed based on surveys on experimental data. Using this model the impact of impulse noise on ATM cell error performance has been traced. Various DSL framings have been found to affect adversely the ATM stream. The interleave depth should be either 1 or large enough to correct the errors. The performance in terms of header and payload errors differs. Time between errored cells exhibits clustering like interarrival times. It is possible within one impulse event to have "good" cells between errored cells. Finally, for higher level applications error free cells is a more appropriate metric than error free seconds.

Linear filter theory based on Wiener filtering is well understood and used widely in many fields of image and signal processing. A field-programmable gate array (FPGA) implementation of a series of nonlinear filters is developed based on the concepts of Volterra series and these are applied to image interpolation problems. More explicitly, the aim is to interpolate one field of a frame of a television picture in real time to form an estimate of the second field. This is known as de-interlacing and is useful in many areas of video processing, for example standards conversion.

Narrowband indoor communication channels can be characterised by statistical distributions. The Rayleigh and Ricean distributions are commonly used to describe measured channels, but in some instances, other distributions may be found to be more appropriate for the task. Determining the reason for one particular distribution being more appropriate for one environment over another may be a non-trivial task, In order to assist in this process, a channel simulation based on the physical structure of the environment is presented, along with narrowband results, obtained both from measurement and simulation experiments. Using these results, the basis for a Nakagami distributed channel, observed in some indoor communication environments, is be shown.

The paper describes a channel simulation designed for the evaluation of a predictive RAKE system, A detailed description of the channel model is given, with reference to the implementation details, and some preliminary analysis of the model outputs is shown, A predictive least squares algorithm is applied to the model, and its performance when training on a noisy signal is shown. The channel used is characterised by a 300 Hz Doppler frequency with an 8 kHz symbol rate. A processing gain of 27 dB (512 chips per symbol) is assumed.